liveBook · Manning

Data transfer

Also called data movement

Data transfer is the movement of digital information between memory locations, computing components, systems, or network endpoints.

How 3 Manning books explain it

Quantization and Fast Inference

Data movement is the transfer of data among storage, memory, and computational locations during computation.

How it is used
Frames data movement as a central efficiency constraint motivating quantization for inference.
How deep it goes
Quantifies energy across memory levels and arithmetic, without detailing transfer mechanisms or runtime implementations.
Examples
Compares approximate energy costs for L1, L2, HBM, and INT8 multiply-add operations.
What you can do after
Understand why memory movement can constrain inference more than arithmetic.

Read how Quantization and Fast Inference explains data transfer →

CUDA for LLMs

Data movement is the movement of data between locations or memory spaces in a computing system.

How it is used
Coordinates CUDA work across separate host and device memory spaces, making explicit transfer part of kernel execution.
How deep it goes
Traces allocation, copies, and kernel execution, then links repeated HBM traffic to attention bottlenecks.
Examples
Uses vector addition and attention matrices to illustrate transfers between host and device and between memory tiers.
What you can do after
Apply CUDA's cudaMalloc, cudaMemcpy, and cudaFree workflow to move vector data through a GPU computation.

Read how CUDA for LLMs explains data transfer →

Fundamentals of Cybersecurity

Data in transit is data being transmitted between systems, devices, or locations over a communication channel.

How it is used
Distinguishes data being transmitted from data at rest and data in use to guide protection choices.
How deep it goes
Defines the state, contrasts its exposure with other data states, and connects transmission risks to controls without implementation steps.
Examples
Uses file transfers, email, instant messages, bank logins, unsecured Wi-Fi eavesdropping, and on-path message alteration as examples.
What you can do after
Evaluate transmission risks and identify encryption or DLP as relevant protections.

Read how Fundamentals of Cybersecurity explains data transfer →

Also covered in

What to know first

Related concepts

Questions about data transfer

What is data transfer?

Data transfer is the movement of digital information between memory locations, computing components, systems, or network endpoints.

Which Manning books explain data transfer?

Quantization and Fast Inference, CUDA for LLMs and Fundamentals of Cybersecurity.

What should I know before learning data transfer?

Manning books that teach data transfer usually introduce encryption, availability zone, float16 and fused multiply-add first.

What is the difference between data transfer and data at rest?

Data transfer is the movement of digital information between memory locations, computing components, systems, or network endpoints. Data at rest is digital information stored on persistent or nonvolatile media rather than actively transmitted across a network.

What kinds of data transfer are there?

Kinds of data transfer include CPU-GPU data transfer.

What is data transfer also called?

Data transfer is also called data movement.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.